New studies of the coolest brown dwarfs are helping astronomers explore the mysterious objects that bridge the gap between stars and planets.

brown dwarf's atmosphere

This artist's illustration shows one concept of a brown dwarf, an object between a regular star and a planet. Some observations suggest that brown dwarfs' atmosphere contain wind-driven cloud structures, reminiscent of weather on planets.

NASA / JPL-Caltech

The distinction between planets and brown dwarfs has blurred considerably over the last few years. The main reason for this was the Wide-field Infrared Survey Explorer (WISE), which launched in December 2009 and scanned the entire sky in the mid-infrared until its retirement in February 2011. (It’s recently been resurrected for asteroid hunting.) One of the key goals of the WISE mission was to identify the coolest brown dwarfs. Just a few years later, WISE images have led to the discovery of several extremely cool brown dwarfs, known as Y dwarfs.

Now, the focus of the astronomers studying these cool neighbors has shifted from discovery and identification to classification and understanding. Two recent papers do just that, and have uncovered some curious properties for these cool brown dwarfs. (Or are they hot planets?)

The first new result is an upcoming Astrophysical Journal paper by Davy Kirkpatrick (Caltech) and his colleagues. Kirkpatrick’s team discovered and characterized one of the coolest objects yet found in the WISE database, WISE J0647–6232. Prior to the late 1990s, astronomers classified all stars based on their spectra into the OBAFGKM categories, with blue O stars being the hottest and red M stars being the coolest. Each subclass is further broken down using 0 through 9, with 0 being the warmest and 9 being the coolest. Thus, a G9 star is just a bit warmer than a K0 star.

But the M class couldn’t contain the rising number of brown dwarfs. The spectral classes L and T were added to help describe and identify these “failed stars,” which lack the sufficient mass and temperature to ignite fusion in their cores. Just two years ago, the Y class was added. Y dwarfs are extremely cool (most have temperatures just at or above room temperature!), making them less than one-tenth the Sun’s temperature but still more than twice Jupiter’s.

Kirkpatrick and his team combined WISE data with ground- and space-based observations spread over three years to carefully study WISE J0647–6232. They determined its distance of about 28 light-years using parallax and estimated its mass, age, and temperature, classifying it as a Y1 dwarf due to its infrared spectral appearance. This makes WISE J0647–6232 one of only a few Y1 dwarfs known and one of the most closely studied to date.

A different study by Trent Dupuy (Harvard-Smithsonian Center for Astrophysics) and Adam Kraus (CfA and University of Texas at Austin) in this week’s Science presents a big-picture view of these objects. Dupuy and Kraus compiled distances for 33 cool T and Y dwarfs using their own Spitzer observations for 16 objects and observations in the literature. The team used its Spitzer-based distances to place the brown dwarfs on a common scale and noticed some interesting patterns.

First, the Y dwarfs are much fainter than T dwarfs—that sounds obvious since they’re cooler, but the brightness dip between them is larger than that between other classes (such as L and T). Second, the potential Y2 in the sample was warmer than the Y0 dwarfs, meaning that something other than just temperature also strongly influences the spectral types of these objects, unlike for L and T dwarfs. Dupuy and Kraus estimated temperatures and masses for most of the Y dwarfs in their sample, finding that they are about 400 Kelvin and about 5 to 20 times the mass of Jupiter.

These objects fit just between the smallest brown dwarfs and the largest planets. The technical boundary between planets and brown dwarfs is about a dozen Jupiter masses, because objects above that limit should intermittently fuse heavy hydrogen. But Y dwarfs make that problematic. Once the James Webb Space Telescope launches with its superb infrared eye, we should have a much better view. The floor for Y dwarfs might end somewhere around objects the mass of Jupiter, and when that happens, we'll have to have another debate on what to call these things!


J.D. Kirkpatrick et al. "Discovery of the Y1 Dwarf WISE J064723.23-623235.5." Astrophysical Journal, in press.

T.J. Dupuy and A.L. Kraus. "Distances, Luminosities, and Temperatures of the Coldest Known Substellar Objects." Science online, 5 September 2013.


Image of Bruce


September 5, 2013 at 6:39 pm

Whoa. Imagine the uproar if the IAU ever decides to upgrade mighty Jupiter from gas GIANT planet to brown DWARF status!

You must be logged in to post a comment.

Image of Larry


September 6, 2013 at 5:55 pm

Quote "Y dwarfs are extremely cool (most have temperatures just at or above room temperature!), making them less than one-tenth the Sun’s temperature but still more than twice Jupiter’s."

Room temp is just 1/10th the Sun's Temp? Then Jupiter is only 1/20th that of the Sun? I think a few zeros are missing?


You must be logged in to post a comment.

Image of Chris Miller

Chris Miller

September 7, 2013 at 4:10 am

Not too many 'zeros' missing. Sun (surface) temperature ~6,000K; room temperature ~300K; Jupiter (surface) temperature ~130K

You must be logged in to post a comment.

Image of Jason


September 8, 2013 at 8:28 pm

Hi Bruce,

Luckily, not to worry. 🙂 A brown dwarf is defined as something in which the core is hot enough to fuse deuterium. This requires temperatures of around 500,000 K, which is over 10 times hotter than Jupiter's core temperature.

You must be logged in to post a comment.

Image of Gibor Basri

Gibor Basri

September 14, 2013 at 12:15 pm

The question raised about whether the Y spectral classification could cross the boundary between planets and brown dwarfs has an easy answer. It could - if the spectral classifiers let it. Since spectral classification is just that, any connection to other physical characteristics is "coincidental". There is no reason objects in a given (somewhat arbitrary, but carefully defined) spectral class can't cross a boundary defined by core deuterium burning (for example) that doesn't affect the particular spectra used for classification. It would be a classification choice (not a clever one, so I'm guessing not likely) to try to cut off the Y spectral class at the nuclear-defined boundary between brown dwarfs and planets.

You must be logged in to post a comment.

You must be logged in to post a comment.